AI-generated code and tests reduce the effectiveness of traditional quality checks, increasing the risk of system failures and liability in production environments.

"When both code and tests are autogenerated, traditional quality checks lose their signaling value, increasing the likelihood of costly failures, outages, and liability exposure in production systems. To responsibly capture productivity gains without undermining trust, organizations must pair AI-generated code with specification-driven development and automated validation techniques that verify real system behavior rather than surface-level compliance." https://protocols-made-fun.com/llms/testing/2026/02/12/second-order-slop.html

"When both code and tests are autogenerated, traditional quality checks lose their signaling value, increasing the likelihood of costly failures, outages, and liability exposure in production systems." protocols-made-fun.com/llms/testing... AI-generated shovels or second...